Seonglae ChoIt performs the diffusion process in latent space rather than pixel space using Latent Diffusion Model. It uses Diffusion Transformer instead of UNet. One H100 GPU can generate up to 5 minutes of video per hour
How to make
Sora Tutorials | OpenAI Academy
Unlock the new opportunities of the AI era by equipping yourself with the knowledge and skills to harness artificial intelligence effectively.
https://academy.openai.com/public/collections/sora-tutorials-2025-03-11
ICML William (Bill) Peebles - TBD
The ICML Logo above may be used on presentations. Right-click and choose
download. It is a vector graphic and may be used at any scale.
https://icml.cc/virtual/2024/39514
Sora: Creating video from text
Sora is an AI model that can create realistic and imaginative scenes from text instructions.
https://openai.com/sora
Factorial Funds | Under The Hood: How OpenAI's Sora Model Works
OpenAI’s Sora model has amazed the world by its ability to generate extremely realistic videos of a wide variety of scenes. Below is a video released by OpenAI that demonstrates the capabilities of the model.
https://www.factorialfunds.com/blog/under-the-hood-how-openai-s-sora-model-works
Vidu
Is It Possible to Get Access to Vidu, the New Chinese Text-to-Video AI Model? - ChatLabs - All-in-one GenAI playground
Explore Vidu AI, a new Chinese text-to-video AI model which considered to be a competitor to Sora, and learn about its features. Discover how to get access to it.
https://writingmate.ai/blog/get-access-to-vidu-ai
